Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Planning Robot Placement for Object Grasping

Manish Saini, Melvin Paul Jacob, Minh Nguyen, Nico Hochgeschwender

TL;DR

The paper tackles mobile robot object grasping by proposing a base-placement planning approach that identifies collision-free and reachability-feasible robot positions around the target object before planning the grasp. It integrates RGB-D sensing and occupancy grid maps within a service-robotics workflow where a user selects the object with hand gestures, and evaluates candidates by a simple navigation/manipulation-cost criterion, avoiding dependence on a precomputed grasp pose. Empirical results on a Toyota HSR show the proposed method achieving 81.7% grasp success, substantially outperforming a baseline that navigates to a fixed location, especially for objects located away from the baseline edge. The work contributes a practical, collision-aware base-placement planner and highlights directions for richer reachability modeling and trajectory-aware collision checks to enhance robustness in cluttered domestic environments.

Abstract

When performing manipulation-based activities such as picking objects, a mobile robot needs to position its base at a location that supports successful execution. To address this problem, prominent approaches typically rely on costly grasp planners to provide grasp poses for a target object, which are then are then analysed to identify the best robot placements for achieving each grasp pose. In this paper, we propose instead to first find robot placements that would not result in collision with the environment and from where picking up the object is feasible, then evaluate them to find the best placement candidate. Our approach takes into account the robot's reachability, as well as RGB-D images and occupancy grid maps of the environment for identifying suitable robot poses. The proposed algorithm is embedded in a service robotic workflow, in which a person points to select the target object for grasping. We evaluate our approach with a series of grasping experiments, against an existing baseline implementation that sends the robot to a fixed navigation goal. The experimental results show how the approach allows the robot to grasp the target object from locations that are very challenging to the baseline implementation.

Planning Robot Placement for Object Grasping

TL;DR

The paper tackles mobile robot object grasping by proposing a base-placement planning approach that identifies collision-free and reachability-feasible robot positions around the target object before planning the grasp. It integrates RGB-D sensing and occupancy grid maps within a service-robotics workflow where a user selects the object with hand gestures, and evaluates candidates by a simple navigation/manipulation-cost criterion, avoiding dependence on a precomputed grasp pose. Empirical results on a Toyota HSR show the proposed method achieving 81.7% grasp success, substantially outperforming a baseline that navigates to a fixed location, especially for objects located away from the baseline edge. The work contributes a practical, collision-aware base-placement planner and highlights directions for richer reachability modeling and trajectory-aware collision checks to enhance robustness in cluttered domestic environments.

Abstract

When performing manipulation-based activities such as picking objects, a mobile robot needs to position its base at a location that supports successful execution. To address this problem, prominent approaches typically rely on costly grasp planners to provide grasp poses for a target object, which are then are then analysed to identify the best robot placements for achieving each grasp pose. In this paper, we propose instead to first find robot placements that would not result in collision with the environment and from where picking up the object is feasible, then evaluate them to find the best placement candidate. Our approach takes into account the robot's reachability, as well as RGB-D images and occupancy grid maps of the environment for identifying suitable robot poses. The proposed algorithm is embedded in a service robotic workflow, in which a person points to select the target object for grasping. We evaluate our approach with a series of grasping experiments, against an existing baseline implementation that sends the robot to a fixed navigation goal. The experimental results show how the approach allows the robot to grasp the target object from locations that are very challenging to the baseline implementation.
Paper Structure (12 sections, 4 figures, 1 table, 2 algorithms)

This paper contains 12 sections, 4 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Approach
  4. Selecting the target object using hand gesture
  5. Planning a robot base placement to pick up the object
  6. Generating robot placement candidates
  7. Checking for risk of collision
  8. Picking up objects based on robot placement candidates
  9. Experimental results and evaluation
  10. Experimental setup and procedure
  11. Results and Discussion
  12. Conclusion and Future Work

Figures (4)

  • Figure 1: Overall execution flow of the base pose planning component.
  • Figure 2: Checking robot placement candidates for risk of collision during the pickup behaviour using RGB-D data. The regions checked for obstacles are visualized by cuboids. Candidates and corresponding cuboids that may result in collision are coloured red, and valid ones green.
  • Figure 3: Experimental Setup
  • Figure 4: Heatmap of approach's success out of 5 trials in each grid.